Energy-aware and real-time service management in cloud computing

Abstract

Performance and energy management in cloud data-center are major concerns that cloud service providers have to encounter due to computing demand has been increasing. Many researchers have proposed solutions to reduce energy consumption such as scheduling and migration while maintaining an appropriate level of performance to customers. In this paper, we propose a new CPU re-allocation algorithm that combined DVFS concept with live migration technique to improve efficiency of energy management and adaptation scheme on real-time service. The proposed algorithm works as three characteristics that are Longest Completion Time (LCT), Highest Utilization (HU) and Lowest Utilization (LU). The result shows the reduction of energy consumption and execution time upon applied characteristics and adaptation scheme.